Retrieval-augmented language models are being increasingly tasked with subjective, contentious, and conflicting queries such as “is aspartame linked to cancer”. To resolve these ambiguous queries, one must search through a large range of websites and consider “which, if any, of this evidence do I find convincing?”. In this work, we study how LLMs answer this question. In particular, we construct ConflictingQA, a dataset that pairs controversial queries with a series of real-world evidence documents that contain different facts (e.g., quantitative results), argument styles (e.g., appeals to authority), and answers (Yes or No). We use this dataset to perform sensitivity and counterfactual analyses to explore which text features most affect LLM predictions. Overall, we find that current models rely heavily on the relevance of a website to the query, while largely ignoring stylistic features that humans find important such as whether a text contains scientific references or is written with a neutral tone. Taken together, these results highlight the importance of RAG corpus quality (e.g., the need to filter misinformation), and possibly even a shift in how LLMs are trained to better align with human judgements.

We present the Berkeley Crossword Solver, a state-of-the-art approach for automatically solving crossword puzzles. Our system works by generating answer candidates for each crossword clue using neural question answering models and then combines loopy belief propagation with local search to find full puzzle solutions. Compared to existing approaches, our system improves exact puzzle accuracy from 57% to 82% on crosswords from The New York Times and obtains 99.9% letter accuracy on themeless puzzles. Our system also won first place at the top human crossword tournament, which marks the first time that a computer program has surpassed human performance at this event. To facilitate research on question answering and crossword solving, we analyze our system’s remaining errors and release a dataset of over six million question-answer pairs.

To create models that are robust across a wide range of test inputs, training datasets should include diverse examples that span numerous phenomena. Dynamic adversarial data collection (DADC), where annotators craft examples that challenge continually improving models, holds promise as an approach for generating such diverse training sets. Prior work has shown that running DADC over 1-3 rounds can help models fix some error types, but it does not necessarily lead to better generalization beyond adversarial test data. We argue that running DADC over many rounds maximizes its training-time benefits, as the different rounds can together cover many of the task-relevant phenomena. We present the first study of longer-term DADC, where we collect 20 rounds of NLI examples for a small set of premise paragraphs, with both adversarial and non-adversarial approaches. Models trained on DADC examples make 26% fewer errors on our expert-curated test set compared to models trained on non-adversarial data. Our analysis shows that DADC yields examples that are more difficult, more lexically and syntactically diverse, and contain fewer annotation artifacts compared to non-adversarial examples.

Prompting language models (LMs) with training examples and task descriptions has been seen as critical to recent successes in few-shot learning. In this work, we show that finetuning LMs in the few-shot setting can considerably reduce the need for prompt engineering. In fact, one can use null prompts, prompts that contain neither task-specific templates nor training examples, and achieve competitive accuracy to manually-tuned prompts across a wide range of tasks. While finetuning LMs does introduce new parameters for each downstream task, we show that this memory overhead can be substantially reduced: finetuning only the bias terms can achieve comparable or better accuracy than standard finetuning while only updating 0.1% of the parameters. All in all, we recommend finetuning LMs for few-shot learning as it is more accurate, robust to different prompts, and can be made nearly as efficient as using frozen LMs.

Adversarial attacks alter NLP model predictions by perturbing test-time inputs. However, it is much less understood whether, and how, predictions can be manipulated with small, concealed changes to the training data. In this work, we develop a new data poisoning attack that allows an adversary to control model predictions whenever a desired trigger phrase is present in the input. For instance, we insert 50 poison examples into a sentiment model’s training set that causes the model to frequently predict Positive whenever the input contains “James Bond”. Crucially, we craft these poison examples using a gradient-based procedure so that they do not mention the trigger phrase. We also apply our poison attack to language modeling (“Apple iPhone” triggers negative generations) and machine translation (“iced coffee” mistranslated as “hot coffee”). We conclude by proposing three defenses that can mitigate our attack at some cost in prediction accuracy or extra human annotation.

Language models (LMs) must be both safe and equitable to be responsibly deployed in practice. With safety in mind, numerous detoxification techniques (e.g., Dathathri et al. 2020; Krause et al. 2020) have been proposed to mitigate toxic LM generations. In this work, we show that these detoxification techniques hurt equity: they decrease the utility of LMs on language used by marginalized groups (e.g., African-American English and minority identity mentions). In particular, we perform automatic and human evaluations of text generation quality when LMs are conditioned on inputs with different dialects and group identifiers. We find that detoxification makes LMs more brittle to distribution shift, especially on language used by marginalized groups. We identify that these failures stem from detoxification methods exploiting spurious correlations in toxicity datasets. Overall, our results highlight the tension between the controllability and distributional robustness of LMs.

Although pretrained Transformers such as BERT achieve high accuracy on in-distribution examples, do they generalize to new distributions? We systematically measure out-of-distribution (OOD) generalization for seven NLP datasets by constructing a new robustness benchmark with realistic distribution shifts. We measure the generalization of previous models including bag-of-words models, ConvNets, and LSTMs, and we show that pretrained Transformers’ performance declines are substantially smaller. Pretrained transformers are also more effective at detecting anomalous or OOD examples, while many previous models are frequently worse than chance. We examine which factors affect robustness, finding that larger models are not necessarily more robust, distillation can be harmful, and more diverse pretraining data can enhance robustness. Finally, we show where future work can improve OOD robustness.

Gradient-based analysis methods, such as saliency map visualizations and adversarial input perturbations, have found widespread use in interpreting neural NLP models due to their simplicity, flexibility, and most importantly, the fact that they directly reflect the model internals. In this paper, however, we demonstrate that the gradients of a model are easily manipulable, and thus bring into question the reliability of gradient-based analyses. In particular, we merge the layers of a target model with a Facade Model that overwhelms the gradients without affecting the predictions. This Facade Model can be trained to have gradients that are misleading and irrelevant to the task, such as focusing only on the stop words in the input. On a variety of NLP tasks (sentiment analysis, NLI, and QA), we show that the merged model effectively fools different analysis tools: saliency maps differ significantly from the original model’s, input reduction keeps more irrelevant input tokens, and adversarial perturbations identify unimportant tokens as being highly important.

Standard test sets for supervised learning evaluate in-distribution generalization. Unfortunately, when a dataset has systematic gaps (e.g., annotation artifacts), these evaluations are misleading: a model can learn simple decision rules that perform well on the test set but do not capture the abilities a dataset is intended to test. We propose a more rigorous annotation paradigm for NLP that helps to close systematic gaps in the test data. In particular, after a dataset is constructed, we recommend that the dataset authors manually perturb the test instances in small but meaningful ways that (typically) change the gold label, creating contrast sets. Contrast sets provide a local view of a model’s decision boundary, which can be used to more accurately evaluate a model’s true linguistic capabilities. We demonstrate the efficacy of contrast sets by creating them for 10 diverse NLP datasets (e.g., DROP reading comprehension, UD parsing, and IMDb sentiment analysis). Although our contrast sets are not explicitly adversarial, model performance is significantly lower on them than on the original test sets—up to 25% in some cases. We release our contrast sets as new evaluation benchmarks and encourage future dataset construction efforts to follow similar annotation processes.

The remarkable success of pretrained language models has motivated the study of what kinds of knowledge these models learn during pretraining. Reformulating tasks as fill-in-the-blanks problems (e.g., cloze tests) is a natural approach for gauging such knowledge, however, its usage is limited by the manual effort and guesswork required to write suitable prompts. To address this, we develop AutoPrompt, an automated method to create prompts for a diverse set of tasks, based on a gradient-guided search. Using AutoPrompt, we show that masked language models (MLMs) have an inherent capability to perform sentiment analysis and natural language inference without additional parameters or finetuning, sometimes achieving performance on par with recent state-of-the-art supervised models. We also show that our prompts elicit more accurate factual knowledge from MLMs than the manually created prompts on the LAMA benchmark, and that MLMs can be used as relation extractors more effectively than supervised relation extraction models. These results demonstrate that automatically generated prompts are a viable parameter-free alternative to existing probing methods, and as pretrained LMs become more sophisticated and capable, potentially a replacement for finetuning.

Adversaries may look to steal or attack black-box NLP systems, either for financial gain or to exploit model errors. One setting of particular interest is machine translation (MT), where models have high commercial value and errors can be costly. We investigate possible exploitations of black-box MT systems and explore a preliminary defense against such threats. We first show that MT systems can be stolen by querying them with monolingual sentences and training models to imitate their outputs. Using simulated experiments, we demonstrate that MT model stealing is possible even when imitation models have different input data or architectures than their target models. Applying these ideas, we train imitation models that reach within 0.6 BLEU of three production MT systems on both high-resource and low-resource language pairs. We then leverage the similarity of our imitation models to transfer adversarial examples to the production systems. We use gradient-based attacks that expose inputs which lead to semantically-incorrect translations, dropped content, and vulgar model outputs. To mitigate these vulnerabilities, we propose a defense that modifies translation outputs in order to misdirect the optimization of imitation models. This defense degrades the adversary’s BLEU score and attack success rate at some cost in the defender’s BLEU and inference speed.

Although neural NLP models are highly expressive and empirically successful, they also systematically fail in counterintuitive ways and are opaque in their decision-making process. This tutorial will provide a background on interpretation techniques, i.e., methods for explaining the predictions of NLP models. We will first situate example-specific interpretations in the context of other ways to understand models (e.g., probing, dataset analyses). Next, we will present a thorough study of example-specific interpretations, including saliency maps, input perturbations (e.g., LIME, input reduction), adversarial attacks, and influence functions. Alongside these descriptions, we will walk through source code that creates and visualizes interpretations for a diverse set of NLP tasks. Finally, we will discuss open problems in the field, e.g., evaluating, extending, and improving interpretation methods.

Multi-hop reading comprehension (RC) questions are challenging because they require reading and reasoning over multiple paragraphs. We argue that it can be difficult to construct large multi-hop RC datasets. For example, even highly compositional questions can be answered with a single hop if they target specific entity types, or the facts needed to answer them are redundant. Our analysis is centered on HotpotQA, where we show that single-hop reasoning can solve much more of the dataset than previously thought. We introduce a single-hop BERT-based RC model that achieves 67 F1—comparable to state-of-the-art multi-hop models. We also design an evaluation setting where humans are not shown all of the necessary paragraphs for the intended multi-hop reasoning but can still answer over 80% of questions. Together with detailed error analysis, these results suggest there should be an increasing focus on the role of evidence in multi-hop reasoning and possibly even a shift towards information retrieval style evaluations with large and diverse evidence collections.

Recent work establishes dataset difficulty and removes annotation artifacts via partial-input baselines (e.g., hypothesis-only model for SNLI or question-only model for VQA). A successful partial-input baseline indicates that the dataset is cheatable. But the converse is not necessarily true: failures of partial-input baselines do not mean the dataset is free of artifacts. We first design artificial datasets to illustrate how the trivial patterns that are only visible in the full input can evade any partial-input baseline. Next, we identify such artifacts in the SNLI dataset—a hypothesis-only model augmented with trivial patterns in the premise can solve 15% of previously-thought “hard” examples. Our work provides a caveat for the use and creation of partial-input baselines for datasets.

Adversarial evaluation stress-tests a model’s understanding of natural language. Because past approaches expose superficial patterns, the resulting adversarial examples are limited in complexity and diversity. We propose human- in-the-loop adversarial generation, where human authors are guided to break models. We aid the authors with interpretations of model predictions through an interactive user interface. We apply this generation framework to a question answering task called Quizbowl, where trivia enthusiasts craft adversarial questions. The resulting questions are validated via live human–computer matches: Although the questions appear ordinary to humans, they systematically stump neural and information retrieval models. The adversarial questions cover diverse phenomena from multi-hop reasoning to entity type distractors, exposing open challenges in robust question answering.

Adversarial examples highlight model vulnerabilities and are useful for evaluation and interpretation. We define universal adversarial triggers: input-agnostic sequences of tokens that trigger a model to produce a specific prediction when concatenated to any input from a dataset. We propose a gradient-guided search over tokens which finds short trigger sequences (e.g., one word for classification and four words for language modeling) that successfully trigger the target prediction. For example, triggers cause SNLI entailment accuracy to drop from 89.94% to 0.55%, 72% of “why” questions in SQuAD to be answered “to kill american people”, and the GPT-2 language model to spew racist output even when conditioned on non-racial contexts. Furthermore, although the triggers are optimized using white-box access to a specific model, they transfer to other models for all tasks we consider. Finally, since triggers are input-agnostic, they provide an analysis of global model behavior. For instance, they confirm that SNLI models exploit dataset biases and help to diagnose heuristics learned by reading comprehension models.

The ability to understand and work with numbers (numeracy) is critical for many complex reasoning tasks. Currently, most NLP models treat numbers in text in the same way as other tokens—they embed them as distributed vectors. Is this enough to capture numeracy? We begin by investigating the numerical reasoning capabilities of a state-of-the-art question answering model on the DROP dataset. We find this model excels on questions that require numerical reasoning, i.e., it already captures numeracy. To understand how this capability emerges, we probe token embedding methods (e.g., BERT, GloVe) on synthetic list maximum, number decoding, and addition tasks. A surprising degree of numeracy is naturally present in standard embeddings. For example, GloVe and word2vec accurately encode magnitude for numbers up to 1,000. Furthermore, character-level embeddings are even more precise—ELMo captures numeracy the best for all pre-trained methods—but BERT, which uses sub-word units, is less exact.

Neural NLP models are increasingly accurate but are imperfect and opaque—they break in counterintuitive ways and leave end users puzzled at their behavior. Model interpretation methods ameliorate this opacity by providing explanations for specific model predictions. Unfortunately, existing interpretation codebases make it difficult to apply these methods to new models and tasks, which hinders adoption for practitioners and burdens interpretability researchers. We introduce AllenNLP Interpret, a flexible framework for interpreting NLP models. The toolkit provides interpretation primitives (e.g., input gradients) for any AllenNLP model and task, a suite of built-in interpretation methods, and a library of front-end visualization components. We demonstrate the toolkit’s flexibility and utility by implementing live demos for five interpretation methods (e.g., saliency maps and adversarial attacks) on a variety of models and tasks (e.g., masked language modeling using BERT and reading comprehension using BiDAF). These demos, alongside our code and tutorials, are available at https://allennlp.org/interpret.

Modern question answering systems have been touted as approaching human performance. However, existing question answering datasets are imperfect tests. Questions are written with humans in mind, not computers, and often do not properly expose model limitations. To address this, we develop an adversarial writing setting, where humans interact with trained models and try to break them. This annotation process yields a challenge set, which despite being easy for trivia players to answer, systematically stumps automated question answering systems. Diagnosing model errors on the evaluation data provides actionable insights to explore in developing robust and generalizable question answering systems.

Local model interpretation methods explain individual predictions by assigning an importance value to each input feature. This value is often determined by measuring the change in confidence when a feature is removed. However, the confidence of neural networks is not a robust measure of model uncertainty. This issue makes reliably judging the importance of the input features difficult. We address this by changing the test-time behavior of neural networks using Deep k-Nearest Neighbors. Without harming text classification accuracy, this algorithm provides a more robust uncertainty metric which we use to generate feature importance values. The resulting interpretations better align with human perception than baseline methods. Finally, we use our interpretation method to analyze model predictions on dataset annotation artifacts.

One way to interpret neural model predictions is to highlight the most important input features—for example, a heatmap visualization over the words in an input sentence. In existing interpretation methods for NLP, a word’s importance is determined by either input perturbation—measuring the decrease in model confidence when that word is removed—or by the gradient with respect to that word. To understand the limitations of these methods, we use input reduction, which iteratively removes the least important word from the input. This exposes pathological behaviors of neural models: the remaining words appear nonsensical to humans and are not the ones determined as important by interpretation methods. As we confirm with human experiments, the reduced examples lack information to support the prediction of any label, but models still make the same predictions with high confidence. To explain these counterintuitive results, we draw connections to adversarial examples and confidence calibration: pathological behaviors reveal difficulties in interpreting neural models trained with maximum likelihood. To mitigate their deficiencies, we fine-tune the models by encouraging high entropy outputs on reduced examples. Fine-tuned models become more interpretable under input reduction, without accuracy loss on regular examples.